It has been said I am a holist. And I am -- to some degree. But I am not an unadulterated holist. I am a modulated holist, a quasi-holist, or, as I prefer to think of it, a sophisticated fellow who believes that both synthesis and dissection play an important role in inference.
A partial proof that what I have just said is true:
P. Tillers, "Webs of Things in the Mind: A New Science of Evidence," 87 Michigan Law Review 1225, 1251-1252 (1989):
[David] Schum sees inference as a network and he believes that networks of inference are extremely intricate. Hence, the webs that Schum weaves around problems of evidence and inference typically consist of many delicate threads, which crisscross in various ways. These threads are sometimes difficult to keep in mind and almost seem to vanish from sight.
Schum's microscopic analyses of evidence and inference may seem unduly intricate; it is natural to wonder whether an entirely different approach to evidence and inference might work better. There has been discussion (although not quite a debate) about the value of fine-woven analyses of evidence. I myself have sometimes wondered if people might do a better job of drawing inferences if, instead of analyzing or dissecting evidence, they would just look at a mass of evidence ‘as a whole,’ try not to think too much about it, and then grunt out a response from somewhere within themselves to the undifferentiated mass of stuff they see in front of them.
This kind of ‘holistic’ alternative to microscopic analysis is practically its own refutation. It is hard even to imagine what it means to take evidence ‘as a whole.’ We perceive slices and various features in almost everything we see—and if we don't, perhaps we can't see anything at all. Moreover, it is hard to imagine how we can imbibe the evidence we ‘see’ without performing some sort of mental analysis, which by definition seems to involve some sort of dissection. In short, it is hard to imagine how we can think holistically even if we want to do so. The admonition not to analyze and dissect almost seems tantamount to advice not to think too carefully about the way you think. One might as well advise you not to think about elephants. You may not have been thinking about elephants before, but once you are told not to do so, you cannot stop thinking about them. Hence, if we are to believe that holistic assessments play a part in inference, we must have a more subtle concept of ‘holism.’ Any theory that assumes an absolute dichotomy between holistic thinking and nonholistic thinking is thoroughly implausible and any theory that admonishes people to think globally rather than locally is vacuous.Cf. P. Tillers, Are There Universal Principles or Forms of Evidential Inference?, in J. Jackson, M. Langer & P. Tillers, eds., Crime, Procedure, and Evidence in a Comparative and International Context (2008).